LV
Casos brutos (1116 municípios)
# A tibble: 6 × 13
year mean sd var min quant25 median quant75 quant90 quant95 max zero zero.perc
<chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <dbl>
1 2001 2.56 11.8 139. 0 0 0 1 4 7 173 645 0.578
2 2002 2.49 12.3 151. 0 0 0 1 4 8 214 702 0.629
3 2003 3.00 15.2 232. 0 0 0 1 4.5 9 303 654 0.586
4 2004 3.38 19.4 376. 0 0 0 1 5 12 464 692 0.620
5 2005 3.45 18.4 338. 0 0 0 1 5.5 12.2 412 662 0.593
6 2006 3.52 17.9 319. 0 0 0 1 5.5 12.2 302 661 0.592
Brasil (whole)
# A tibble: 5 × 13
year mean sd var min quant25 median quant75 max zero zero.perc `NA` NA.perc
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <dbl> <int> <dbl>
1 2001 1.05e-5 5.52e-5 3.05e-9 0 0 0 5.87e-5 0.00120 5089 0.913 12 0.00215
2 2002 8.95e-6 5.44e-5 2.96e-9 0 0 0 4.12e-5 0.00128 5146 0.924 12 0.00215
3 2003 1.09e-5 6.11e-5 3.73e-9 0 0 0 5.66e-5 0.00167 5098 0.915 12 0.00215
4 2004 9.90e-6 5.74e-5 3.29e-9 0 0 0 4.83e-5 0.00173 5140 0.922 8 0.00144
5 2005 1.04e-5 5.67e-5 3.22e-9 0 0 0 5.62e-5 0.00131 5110 0.917 8 0.00144
LTA
Casos brutos (3212 municípios)
# A tibble: 6 × 13
year mean sd var min quant25 median quant75 quant90 quant95 max zero zero.perc
<chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <dbl>
1 2001 10.6 45.0 2025. 0 0 1 6 23 50 1772 1289 0.401
2 2002 9.66 33.8 1141. 0 0 1 6 23 49.4 1144 1224 0.381
3 2003 10.1 46.9 2196. 0 0 1 6 24 49 2107 1294 0.403
4 2004 9.28 34.9 1215. 0 0 1 5 20 43 1023 1402 0.436
5 2005 8.62 29.5 869. 0 0 1 5 20 44.4 928 1271 0.396
6 2006 6.93 23.9 570. 0 0 1 4 16 34 720 1378 0.429
Brasil (whole)
# A tibble: 5 × 13
year mean sd var min quant25 median quant75 max zero zero.perc `NA` NA.perc
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <dbl> <int> <dbl>
1 2001 0.000293 0.00122 1.49e-6 0 0 0 0.00157 0.0416 3637 0.653 12 0.00215
2 2002 0.000284 0.00104 1.09e-6 0 0 0 0.00151 0.0225 3572 0.641 12 0.00215
3 2003 0.000287 0.00107 1.16e-6 0 0 0 0.00160 0.0287 3642 0.654 12 0.00215
4 2004 0.000242 0.000924 8.53e-7 0 0 0 0.00130 0.0201 3754 0.674 8 0.00144
5 2005 0.000241 0.000851 7.24e-7 0 0 0 0.00132 0.0149 3623 0.650 8 0.00144
UR
# A tibble: 5 × 9
year mean sd var min quant25 median quant75 max
<int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 2001 74.5 7.94 63.1 59.3 67.4 74.6 81.2 90.1
2 2002 74.5 7.67 58.9 59.0 67.7 74.4 81.1 90.0
3 2003 74.6 7.96 63.4 59.2 67.7 74.6 81.5 90.1
4 2004 74.6 7.79 60.7 58.8 67.8 74.6 81.3 89.6
5 2005 73.6 8.21 67.4 58.7 66.5 73.8 80.5 89.8
SDII
# A tibble: 5 × 9
year mean sd var min quant25 median quant75 max
<int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 2001 4.65 0.855 0.730 2.4 4.18 4.59 5.03 8.72
2 2002 4.46 0.797 0.636 2.48 4.02 4.42 4.82 8.98
3 2003 4.64 0.787 0.619 2.45 4.22 4.57 4.99 9.08
4 2004 4.69 0.738 0.545 2.82 4.27 4.58 5.04 8.68
5 2005 4.62 0.823 0.677 2.6 4.12 4.47 5 9.48
TASMAX
# A tibble: 5 × 9
year mean sd var min quant25 median quant75 max
<int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 2001 27.5 2.89 8.33 20.1 25.4 27.4 29.8 33.8
2 2002 27.5 2.98 8.89 20.2 25.1 27.4 29.9 33.9
3 2003 27.6 2.93 8.61 20.3 25.3 27.5 30.0 34.1
4 2004 27.5 3.06 9.37 19.8 25.2 27.4 29.8 33.9
5 2005 28.0 2.93 8.60 20.3 25.8 27.8 30.4 33.9
TASMIN
# A tibble: 5 × 9
year mean sd var min quant25 median quant75 max
<int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 2001 17.7 2.97 8.85 10.9 15.2 17.8 20.2 25.4
2 2002 17.7 3.00 8.99 10.9 15.2 17.8 20.2 25.4
3 2003 17.9 2.93 8.61 11.2 15.4 18.0 20.3 25.4
4 2004 17.8 3.05 9.33 10.9 15.2 17.9 20.3 25.6
5 2005 17.9 2.96 8.78 11.1 15.5 18.1 20.4 25.6
PREC_MEAN
# A tibble: 5 × 9
year mean sd var min quant25 median quant75 max
<int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 2001 4.42 0.914 0.836 1.99 3.93 4.27 4.86 8.72
2 2002 4.21 0.847 0.717 2.23 3.74 4.08 4.60 8.96
3 2003 4.40 0.854 0.729 2.16 3.90 4.26 4.81 8.98
4 2004 4.44 0.811 0.658 2.52 3.95 4.28 4.86 8.65
5 2005 4.37 0.889 0.790 2.34 3.84 4.17 4.83 9.40
PREC_SUM
# A tibble: 5 × 9
year mean sd var min quant25 median quant75 max
<int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 2001 134. 27.7 770. 60.2 119. 129. 147. 264.
2 2002 128. 25.7 662. 67.2 113. 124. 139. 271.
3 2003 133. 25.9 673. 65.0 118. 129. 146. 272.
4 2004 135. 24.7 609. 77.0 120. 130. 148. 264.
5 2005 133. 27.0 727. 70.7 116. 126. 147. 285.
Modelagem
PREC_MEANePREC_SUMsão “super” colineares / dependentes / redundantes / dizem a mesma coisa. Usaremos apenas uma, aPREC_SUM.Como estamos trabalhando com incidências, não contagens absolutas, modelos com inflação de zero não são viáveis. Pela presença de zeros, distribuições não-simétricas para dados contínuos, como a Gama e a Normal Inversa, também não são viáveis. Sendo assim, ficamos com a distribuição Normal.
Ajustamos modelos mistos / multiníveis / hierárquicos à nível de município, para assim acomodar a dependência das cinco observações de cada. Tentamos também modelar essa dependência intra-município de um modo temporal, mas não obtivemos convergência numérica. Sendo assim, acomodamos o efeito dos anos no efeito fixo, junto das variáveis climáticas.
Para as variáveis estatisticamente significativas dentro de cada bioma, gráficos são fornecidos para a partir deles limiares serem obtidos.
Modelando LV por bioma
Bioma Amazonia + Pantanal
# A tibble: 5 × 12
year mean sd var min quant25 median quant75 max n zero zero.perc
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <int> <dbl>
1 2001 0.00000954 0.0000520 2.71e-9 0 0 0 0 5.84e-4 554 508 0.917
2 2002 0.0000102 0.0000625 3.91e-9 0 0 0 0 8.88e-4 554 508 0.917
3 2003 0.0000125 0.0000703 4.94e-9 0 0 0 0 8.08e-4 554 511 0.922
4 2004 0.0000142 0.0000711 5.05e-9 0 0 0 0 6.32e-4 554 503 0.908
5 2005 0.0000192 0.000102 1.04e-8 0 0 0 0 1.31e-3 554 501 0.904
# A tibble: 7 × 4
var Estimate `Std. Error` p.value
<chr> <dbl> <dbl> <dbl>
1 (Intercept) -0.00462 0.00122 NA
2 year 0.00000229 0.000000611 0.000185
3 UR 0.000000115 0.000000212 0.589
4 SDII 0.000000865 0.00000121 0.475
5 TASMAX 0.000000862 0.000000906 0.341
6 TASMIN 0.000000606 0.000000863 0.483
7 PREC_SUM 0.0000000125 0.0000000380 0.741 
Bioma Caatinga
# A tibble: 5 × 12
year mean sd var min quant25 median quant75 max n zero zero.perc
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <int> <dbl>
1 2001 0.0000273 0.0000830 0.00000000688 0 0 0 0 1.13e-3 1079 847 0.785
2 2002 0.0000193 0.0000617 0.00000000381 0 0 0 0 6.76e-4 1079 884 0.819
3 2003 0.0000243 0.0000823 0.00000000678 0 0 0 0 1.24e-3 1079 873 0.809
4 2004 0.0000232 0.0000791 0.00000000626 0 0 0 0 9.94e-4 1079 888 0.823
5 2005 0.0000261 0.0000800 0.00000000640 0 0 0 0 1.11e-3 1079 875 0.811
# A tibble: 7 × 4
var Estimate `Std. Error` p.value
<chr> <dbl> <dbl> <dbl>
1 (Intercept) -0.00191 0.00122 NA
2 year 0.000000925 0.000000607 0.128
3 UR 0.000000729 0.000000186 0.0000924
4 SDII -0.00000621 0.00000136 0.00000529
5 TASMAX 0.00000153 0.000000693 0.0269
6 TASMIN 0.00000171 0.000000612 0.00520
7 PREC_SUM -0.000000123 0.0000000479 0.0105 
Bioma Cerrado
# A tibble: 5 × 12
year mean sd var min quant25 median quant75 max n zero zero.perc
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <int> <dbl>
1 2001 0.0000148 0.0000724 0.00000000525 0 0 0 0 1.20e-3 1070 965 0.902
2 2002 0.0000160 0.0000891 0.00000000795 0 0 0 0 1.28e-3 1070 970 0.907
3 2003 0.0000190 0.0000875 0.00000000766 0 0 0 0 1.67e-3 1070 931 0.870
4 2004 0.0000141 0.0000650 0.00000000423 0 0 0 0 8.63e-4 1070 962 0.899
5 2005 0.0000119 0.0000507 0.00000000257 0 0 0 0 7.46e-4 1070 956 0.893
# A tibble: 7 × 4
var Estimate `Std. Error` p.value
<chr> <dbl> <dbl> <dbl>
1 (Intercept) 0.00274 0.00106 NA
2 year -0.00000140 0.000000530 0.00823
3 UR 0.000000332 0.000000178 0.0629
4 SDII 0.00000178 0.00000115 0.122
5 TASMAX 0.000000890 0.000000639 0.164
6 TASMIN 0.000000869 0.000000551 0.114
7 PREC_SUM 0.0000000422 0.0000000393 0.282 
Bioma Mata Atlantica
# A tibble: 5 × 12
year mean sd var min quant25 median quant75 max n zero zero.perc
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <int> <dbl>
1 2001 0.00000291 0.0000263 6.93e-10 0 0 0 0 0.000890 2751 2654 0.965
2 2002 0.00000213 0.0000220 4.85e-10 0 0 0 0 0.000598 2751 2669 0.970
3 2003 0.00000247 0.0000265 7.05e-10 0 0 0 0 0.000678 2751 2669 0.970
4 2004 0.00000254 0.0000375 1.40e- 9 0 0 0 0 0.00173 2751 2672 0.971
5 2005 0.00000223 0.0000267 7.14e-10 0 0 0 0 0.00112 2751 2663 0.968
# A tibble: 7 × 4
var Estimate `Std. Error` p.value
<chr> <dbl> <dbl> <dbl>
1 (Intercept) 4.27e-4 0.000289 NA
2 year -2.12e-7 0.000000145 0.144
3 UR 6.91e-9 0.0000000472 0.884
4 SDII -1.16e-6 0.000000357 0.00117
5 TASMAX 2.10e-8 0.000000160 0.896
6 TASMIN 2.88e-7 0.000000126 0.0219
7 PREC_SUM -8.46e-9 0.0000000105 0.419 
Bioma Pampa
# A tibble: 5 × 12
year mean sd var min quant25 median quant75 max n zero zero.perc
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <int> <dbl>
1 2001 0.00000183 0.0000197 3.88e-10 0 0 0 0 0.000212 118 115 0.975
2 2002 0.00000274 0.0000296 8.73e-10 0 0 0 0 0.000318 118 115 0.975
3 2003 0.00000367 0.0000394 1.55e- 9 0 0 0 0 0.000425 118 114 0.966
4 2004 0.00000137 0.0000148 2.19e-10 0 0 0 0 0.000159 118 115 0.975
5 2005 0.00000229 0.0000247 6.08e-10 0 0 0 0 0.000266 118 115 0.975
# A tibble: 7 × 4
var Estimate `Std. Error` p.value
<chr> <dbl> <dbl> <dbl>
1 (Intercept) 0.000231 0.000667 NA
2 year -0.000000146 0.000000336 0.665
3 UR 0.000000550 0.000000187 0.00428
4 SDII -0.00000361 0.00000135 0.00870
5 TASMAX 0.00000156 0.000000541 0.00493
6 TASMIN -0.000000141 0.000000314 0.655
7 PREC_SUM 0.0000000139 0.0000000326 0.671 
Modelando LTA por bioma
Bioma Amazonia + Pantanal
# A tibble: 5 × 12
year mean sd var min quant25 median quant75 max n zero zero.perc
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <int> <dbl>
1 2001 0.000364 0.00102 0.00000104 0 0 0.00000772 0.000252 0.0128 554 268 0.484
2 2002 0.000542 0.00147 0.00000215 0 0 0.00000426 0.000362 0.0138 554 275 0.496
3 2003 0.000883 0.00212 0.00000449 0 0 0.0000315 0.000815 0.0287 554 255 0.460
4 2004 0.000918 0.00201 0.00000405 0 0 0.0000772 0.000820 0.0191 554 234 0.422
5 2005 0.000792 0.00171 0.00000294 0 0 0.0000744 0.000617 0.0149 554 228 0.412
# A tibble: 7 × 4
var Estimate `Std. Error` p.value
<chr> <dbl> <dbl> <dbl>
1 (Intercept) -0.241 0.0318 NA
2 year 0.000119 0.0000159 8.49e-14
3 UR 0.0000123 0.00000537 2.23e- 2
4 SDII 0.0000330 0.0000308 2.85e- 1
5 TASMAX 0.0000246 0.0000226 2.77e- 1
6 TASMIN 0.0000338 0.0000218 1.22e- 1
7 PREC_SUM -0.00000114 0.000000976 2.41e- 1
Bioma Caatinga
# A tibble: 5 × 12
year mean sd var min quant25 median quant75 max n zero zero.perc
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <int> <dbl>
1 2001 0.000752 0.00201 0.00000403 0 0 0.0000598 0.000746 0.0416 1079 506 0.469
2 2002 0.000683 0.00148 0.00000220 0 0 0.0000561 0.000722 0.0180 1079 485 0.449
3 2003 0.000435 0.00119 0.00000142 0 0 0 0.000366 0.0172 1079 566 0.525
4 2004 0.000384 0.00116 0.00000135 0 0 0 0.000323 0.0201 1079 561 0.520
5 2005 0.000350 0.000825 0.000000681 0 0 0 0.000342 0.0128 1079 543 0.503
# A tibble: 7 × 4
var Estimate `Std. Error` p.value
<chr> <dbl> <dbl> <dbl>
1 (Intercept) 0.207 0.0224 NA
2 year -0.000103 0.0000112 2.63e-20
3 UR 0.0000106 0.00000340 1.84e- 3
4 SDII -0.000115 0.0000249 3.85e- 6
5 TASMAX 0.00000606 0.0000127 6.33e- 1
6 TASMIN 0.0000292 0.0000112 8.96e- 3
7 PREC_SUM -0.000000212 0.000000868 8.07e- 1
Bioma Cerrado
# A tibble: 5 × 12
year mean sd var min quant25 median quant75 max n zero zero.perc
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <int> <dbl>
1 2001 0.000244 0.00112 0.00000125 0 0 0 0.000136 0.0224 1070 588 0.550
2 2002 0.000244 0.00100 0.00000101 0 0 0 0.000100 0.0169 1070 599 0.560
3 2003 0.000255 0.000856 0.000000732 0 0 0 0.000131 0.0135 1070 625 0.584
4 2004 0.000244 0.000839 0.000000703 0 0 0 0.000143 0.0129 1070 611 0.571
5 2005 0.000238 0.000771 0.000000595 0 0 0 0.000154 0.0110 1070 600 0.561
# A tibble: 7 × 4
var Estimate `Std. Error` p.value
<chr> <dbl> <dbl> <dbl>
1 (Intercept) 0.0160 0.0122 NA
2 year -0.00000841 0.00000614 0.171
3 UR 0.00000803 0.00000211 0.000143
4 SDII -0.0000241 0.0000134 0.0736
5 TASMAX 0.00000963 0.00000748 0.198
6 TASMIN 0.0000107 0.00000643 0.0966
7 PREC_SUM 0.00000101 0.000000460 0.0288 
Bioma Mata Atlantica
# A tibble: 5 × 12
year mean sd var min quant25 median quant75 max n zero zero.perc
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <int> <dbl>
1 2001 0.000144 0.000736 0.000000541 0 0 0 0 0.0182 2751 2074 0.754
2 2002 0.000118 0.000590 0.000000348 0 0 0 0 0.0112 2751 2126 0.773
3 2003 0.000105 0.000802 0.000000643 0 0 0 0 0.0255 2751 2163 0.786
4 2004 0.0000806 0.000388 0.000000151 0 0 0 0 0.0109 2751 2166 0.787
5 2005 0.0000768 0.000331 0.000000110 0 0 0 0 0.00859 2751 2182 0.793
# A tibble: 7 × 4
var Estimate `Std. Error` p.value
<chr> <dbl> <dbl> <dbl>
1 (Intercept) 0.0534 0.00679 NA
2 year -0.0000270 0.00000340 1.99e-15
3 UR 0.00000572 0.00000105 5.75e- 8
4 SDII -0.0000226 0.00000815 5.52e- 3
5 TASMAX 0.0000189 0.00000360 1.65e- 7
6 TASMIN 0.00000592 0.00000287 3.95e- 2
7 PREC_SUM -0.000000543 0.000000237 2.22e- 2
Bioma Pampa
# A tibble: 5 × 12
year mean sd var min quant25 median quant75 max n zero zero.perc
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> <int> <dbl>
1 2001 0.000000935 0.00000896 8.02e-11 0 0 0 0 0.0000957 118 114 0.966
2 2002 0.00000198 0.0000127 1.60e-10 0 0 0 0 0.000105 118 114 0.966
3 2003 0.00000490 0.0000523 2.73e- 9 0 0 0 0 0.000568 118 116 0.983
4 2004 0.00000334 0.0000258 6.65e-10 0 0 0 0 0.000265 118 115 0.975
5 2005 0.00000135 0.00000898 8.07e-11 0 0 0 0 0.0000781 118 114 0.966
# A tibble: 7 × 4
var Estimate `Std. Error` p.value
<chr> <dbl> <dbl> <dbl>
1 (Intercept) 0.000911 0.00179 NA
2 year -0.000000495 0.000000898 0.581
3 UR -0.000000303 0.000000423 0.475
4 SDII 0.00000410 0.00000311 0.188
5 TASMAX 0.00000190 0.00000120 0.114
6 TASMIN 0.000000356 0.000000739 0.630
7 PREC_SUM 0.000000296 0.0000000781 0.000166
Referência
A análise estatística foi realizada no ambiente de computação estatística
R(R Core Team, 2022). Os principais pacotesRutilizados foram o {dplyr} (Wickham et al., 2022), {tidyr} (Wickham & Girlich, 2022), {ggplot2} (Wickham, 2016), {patckwork} (Pedersen, 2020), {geobr} (Pereira & Gonçalves, 2022), {rlang} (Henry & Wickham, 2022), {purrr} (Henry & Wickham, 2020), {psych} (Revelle, 2022), {lme4} (Bates et al., 2015), e {lmerTest} (Kuznetsova et al., 2017).
R Core Team (2022). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/
Wickham, H., François, R., Henry, L., Müller, K. (2022). dplyr: A Grammar of Data Manipulation. R package version 1.0.9. https://CRAN.R-project.org/package=dplyr
Wickham, H., Girlich, M. (2022). tidyr: Tidy Messy Data. R package version 1.2.0, https://CRAN.R-project.org/package=tidyr
Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York
Pedersen, T. L. (2020). patchwork: The Composer of Plots. R package version 1.1.1. https://CRAN.R-project.org/package=patchwork
Pereira, R. H. M., Gonçalves, C. N. (2022). geobr: Download Official Spatial Data Sets of Brazil. R package version 1.6.5999, <https://github.com/ipeaGIT/geobr
Henry, L., Wickham, H. (2022). rlang: Functions for Base Types and Core R and ‘Tidyverse’ Features. R package version 1.0.3, https://CRAN.R-project.org/package=rlang
Henry, L., Wickham, H. (2020). purrr: Functional Programming Tools. R package version 0.3.4, https://CRAN.R-project.org/package=purrr
Revelle, W. (2022). psych: Procedures for Personality and Psychological Research, Northwestern University, Evanston, Illinois, USA, R package version 2.2.3, https://CRAN.R-project.org/package=psych
Bates, D., Maechler, M., Bolker, B., Walker, S. (2015). Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software, 67(1), 1-48. doi:10.18637/jss.v067.i01
Kuznetsova, A., Brockhoff, P. B., Christensen, R. H. B. (2017). lmerTest Package: Tests in Linear Mixed Effects Models. Journal of Statistical Software, 82(13), 1-26. doi:10.18637/jss.v082.i13